<<

Lunar and Planetary Science XXXIV (2003) 1824.pdf

DICHOTOMY BOUNDARY AT , : DEVELOPED IN A SEDIMENTARY DEPOSIT. R. P. Irwin III1,2, T. R. Watters1, A. D. Howard2, T. A. Maxwell1, and R. A. Craddock1, 1Center for Earth and Planetary Studies, National Air and Space Museum, Smithsonian Institution, 4th St. and Inde- pendence Ave. SW, Washington DC 20560-0315, [email protected], [email protected], [email protected], [email protected]. 2Dept. Environmental Sciences, University of Virginia, Charlottesville VA 22903, [email protected].

Introduction: The origin of fretted and knobby ter- rain on Mars has remained uncertain since the land- forms were first described by Sharp in 1973 [1]. Sub- sequent studies have focused primarily on the fretted terrain in northern , where investigators generally agree that ground ice has been important in modifying knobs and fretted valleys [2-5]. The initial processes isolating mesas from the high-standing terrain are less certain. Some fretted channels exhibit charac- teristics that suggest origin by fluvial erosion, despite their poorly developed drainage networks [5]. Other proposed mechanisms include crustal extension and structural control of sapping [6]. Situated near the equator at the crustal di- chotomy boundary, Aeolis Mensae (Fig. 1) provides a pristine example of fretted terrain development without the younger attributed to ice. We examined o o o o an area bounded by 10 S, 10 N, 120 E, and 150 E, ad- jacent to the cratered and dissected area described by Fig. 1. Mars Orbiter Laser Altimeter contour map of Irwin and Howard [7]. Here we present evidence for a part of the study area in eastern Aeolis Mensae, compositional difference between the Aeolis Mensae bounded by 2oN, 15oS, 135oE, and 150oE, 889 km and the adjacent highland , and we discuss the in- across. The fretted terrain mesa materials appear to be teraction of fluvial networks with the dichotomy emplaced at a higher topographic level than the termini boundary in this region. Our observations indicate that of wide theater-headed valleys. crater (left-center) Aeolis Mensae fretted terrain developed in a thick contains thick sediments, higher than the crater rim, that sedimentary deposit. Sedimentary layers were em- may be of similar origin to layers in the fretted unit. placed and eroded as fluvial activity declined, with Locations of Figs. 2 and 3 are indicated by arrows. minimal influence from highland valley networks. Distinguishing highland from sedimen- Relationships with highland valley networks: tary deposits: The extensive degradation of Aeolis The dichotomy boundary in the study area consists of a Mensae relative to the adjacent cratered highlands, steep scarp dropping into elongated open and closed where 20–50 m deep valley networks are preserved [7], basins (Fig. 1). The density of valley networks on this suggests either that the mensae are composed of differ- north-facing scarp is low relative to degraded crater ent materials than the highland crust, or that geomor- rims in the adjacent highlands. The few valley networks phic proceses were concentrated along the dichotomy crossing the boundary commonly steepen abruptly near boundary [1]. Malin and Edgett [8] describe criteria for the scarp [7], consistent with a base-level reduction or distinguishing sedimentary deposits on Mars from high- southward advancement of the boundary scarp [7,9]. land bedrock. These criteria include step-like expo- These valley networks do not appear to continue into sures of horizontal sedimentary layers, fluting or yar- the multibasinal lowland knobby terrain. dangs, and an absence of boulders in talus slopes. The Aeolis Mensae materials appear to stratigraphi- Where thick sequences occur, a massive (or perhaps cally overlie or “dam” highland valley networks that layered at sub-meter resolution) deposit overlies the drain toward the dichotomy boundary (Fig. 1). The thinner stair-stepped sequences. mesas and knobs are not dissected by valley networks as (MOC) images suggest that knobs and mesas in Aeolis are slopes in the adjacent highlands (Fig. 2). It is there- Mensae include the same and landforms as fore likely that erosion of Aeolis Mensae occurred in sedimentary deposits in other regions on Mars [8] (Fig. the absence of abundant , during a time 2). when activity of highland valley networks was waning. Lunar and Planetary Science XXXIV (2003) 1824.pdf

AEOLIS MENSAE FRETTED TERRAIN IN SEDIMENTARY DEPOSITS: R. P. Irwin et al.

Very limited fluvial activity is that layered deposits in Terra Merid- continued after material was iani, Arabia Terra, , and stripped to form the knobby along the dichotomy boundary may share and fretted terrain. Crater a common origin, as fine-grained sedi- counts [11,12] and the activ- mentary deposits were stripped from ity of highland drainage ba- low-standing settings and transported to sins can be used to constrain their current locations by wind. the emplacement and erosion of the Aeolis Mensae to the Fig. 2 (left). MOC image E05-01183 late and (3 km across) of mesas in Aeolis Men- periods. sae. Mass wasted deposits on the mesa Possible emplacement slopes (bottom) are free of large boulders and erosional processes: even at full resolution (5.8 m/pixel). Sedimentary materials could Between mesas a layered deposit has have been transported into the been partially stripped in the upper part Aeolis Mensae region from of the image. An indurated layer caps the highlands by fluvial activ- the mesa. ity, or they could have origi- nated from either the high- Fig. 3 (right). Subframe of Mars lands or lowlands as an airfall Odyssey THEMIS daytime infrared im- deposit. For their present I00957001 (32 km across). Aeolis appearance and erosion, the Mensae fretted terrain (top) formed in an mesas must have disaggre- etched sedimentary layer (center), which gated entirely to grain sizes overlies cratered terrain and valley net- that are transportable by works. A higher-inertia layer is exposed wind, so emplacement between knobs at the top of the image. mechanisms should be consis- In MOC images, this high-inertia mate- tent with this sorting require- rial underlies a largely stripped, thin ment. layer, which itself underlies the thick A more durable cap rock mesa-forming unit. in many flat-topped mesas (Fig. 2). Erosion of References: [1] Sharp R. P. (1973) JGR, these knobs may have been 78, 4073–4083. [2] Squyres S. W. limited by the backwasting or (1978) Icarus, 34, 600–613. [3] Lucchita disaggregation of the capping B. K. (1984) JGR, 89, B409–B418. [4] unit. Transport of disaggre- McGill G. E. (2000) JGR, 105, 6945- gated sedimentary materials 6959. [5] Carr M. H. (2001) JGR, 106, into the lowlands could have 23,751–23,593. [6] McGill G. E. and been accomplished by wind, Dimitriou A. M. (1990) JGR, 95,12,595– water, or ice. Yardangs, 12,605, 1990. [7] Irwin R. P. and How- etched terrain, absence of ard A. D. (2002) JGR, continuous flow paths, and 10.1029/2001JE001818. [8] Malin M. C. the temporal relationships and Edgett K. S. (2000) Science, 290, with valley networks favor 1927-1937. [9] Kochel R. C. and Peake wind as the transport medium. R. T. (1984) JGR, 89, C336–C350. [10] Regional and planetary Malin M. C. and Edgett K. S. (2001) correlations: Malin and JGR, 106, 23,429–23,570. [11] Frey, H. Edgett [8,10] described sedi- V. et al. (1988) Proc. LPSC 18, 679– mentary mantles covering 699. [12] Maxwell T. A. and McGill G. substantial areas of low- E. (1988) Proc. LPSC 18, 701–711. standing ground. These lay- ered units showed no obvious relationship to highland val- ley networks. One possibility